Workbook on Digital Private Papers > Administrative and preservation metadata > Persistent identifiers
Persistent identifiers
Identifiers and the Fedora repository software
The Fedora repository (used by Paradigm in its protoype preservation repository, see Chapter 07 Digital repositories) assigns identifiers to all the objects ingested into it; these identifiers are unique within the local Fedora repository and may suffice for a dark archive. If the material is to be made available in another namespace, additional identifiers may be necessary. Further information, including rules for the formation of Fedora identifiers, is available at http://www.fedora.info/definitions/identifiers/.
Fedora digital objects
The Fedora digital object model represents each object as a container which includes the following basic components:
- PID - a unique identifier for the object and its container.
- Object Properties - key object metadata to enable management of the object within the repository.
- Datastreams - one or more to hold the content or metadata items.
- Disseminators - optional pointers to methods to allow runtime representations to be generated.
Fedora identifier syntax
Fedora identifiers take the following form:
[PID] / [Component Name]
where PID is of the form [Namespace Identifier] : [Object Identifier]
Hypothetical example(s):
aNamespace:Object23 - unique identifier for the object (PID)
aNamespace:Object23/DC - component (Dublin Core metadata datastream)
aNamespace:Object23/file1 - component (content datastream)
Namespace Identifier The namespace in which the object identifier is unique. For example:
- aNamespace.
Object Identifier A unique string for the object in the namespace. For example:
- Object23.
Component Name A string which identifies a component of the object, this references either:
- A datastream containing:
- Metadata - aNamespace:Object23/DC.
- Or a digital file - aNamespace:Object-23/file1.
- Or a method-call to access a run-time service:
- aNamespace:Object-23/service-tobecalled.
